This item is only available as the following downloads:
1 DETECTION OF PAV1 INFECTED CONSPECIFICS AND ITS POPULATION STRUCTURING DYNAMICS IN THE CARIBBEAN SPINY LOBSTER ( PANULIRUS ARGUS ) By JOSHUA RUSS ANDERSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Joshua Russ Anderson
3 To my parents
4 ACKNOWLEDGMENTS I thank Dr. Donald Behringer for his mentorship, knowledge, and support. I thank Dr. Tom Frazer and Dr. Andrew Kane for serving on my graduate committee and all the participants in data collection, especially Jason Spadaro. Finally, I thank my parents for all of their love and support.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TAB LES ................................ ................................ ................................ ........................... 6 LIST OF FIGURES ................................ ................................ ................................ ......................... 7 LIST OF ABBREVIATIONS ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 I NTRODUCTION ................................ ................................ ................................ .................. 11 2 M ETHODS ................................ ................................ ................................ ............................. 19 Mode of PaV1 infected Conspecific Detection and Cue Source ................................ ............ 19 Experimental Lobsters ................................ ................................ ................................ ..... 19 Experimental Setup ................................ ................................ ................................ ......... 19 Chemical Detection of PaV1 infected Conspecifics ................................ ....................... 20 PaV1 infected Conspecific Urine Detection ................................ ................................ ... 21 Visual Cues in PaV1 Detection ................................ ................................ ....................... 22 Data Analyses ................................ ................................ ................................ .................. 22 PaV1 infected Conspecific Avoidance Range ................................ ................................ ........ 23 Experimental Setup and Procedure ................................ ................................ ................. 23 Data Analysis ................................ ................................ ................................ ................... 24 Influence of PaV1 Avoidance on Juvenile Population Structuring Dynamics ....................... 24 Experimental Setup ................................ ................................ ................................ ......... 24 Experimental Procedure ................................ ................................ ................................ .. 25 Data Analysis ................................ ................................ ................................ ................... 26 3 RESULTS ................................ ................................ ................................ ............................... 32 Mode of PaV1 infected Conspecific Detection and Cue Source ................................ ............ 32 PaV1 infected Conspecific Avoidance Range ................................ ................................ ........ 32 Influence of PaV1 Avoidance on Juvenile Population Structuring Dynamics ....................... 33 4 DISCUSSION ................................ ................................ ................................ ......................... 44 LIST OF REFERENCES ................................ ................................ ................................ ............... 52 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 59
6 LIST OF TABLES Table page 3 1 Results determining the detectio n of PaV1 infected conspecifics ................................ ..... 36 3 2 Results used to determine the range of t he PaV1 olfactory avoidance cue ....................... 36 3 3 Results used to determine effect of disease avoidance on the number of lobsters per site over time and under high and low flow regimes ................................ ......................... 37 3 4 Results used to determine the spatial structuring capabilities of PaV1 relative to the number of lobsters on the site ................................ ................................ ............................ 37 3 5 Results used to determine the spatial structuring capabilities of PaV1 relative to the number of lobsters on the site ................................ ................................ ............................ 38 3 6 Results used to determine the influence of disease avoidance on the juvenile populations structure ................................ ................................ ................................ .......... 38 3 7 Results used to determine the affect of PaV1 infected conspecific avoidance on lobster movement ................................ ................................ ................................ ............... 39 3 8 Results used to determine the affect of PaV1 infected conspecific avoidance on lobster movement ................................ ................................ ................................ ............... 39
7 LIST OF FIGURES Figure page 2 1 Florida Bay, FL (USA), the location of lobst er collection and study sites ........................ 28 2 2 Y maze design used to determine the mode of PaV1 infected conspecific dete ction and the source of the cue ................................ ................................ ................................ .... 28 2 3 Y maze set up for experiments used to determine if detection of PaV1 infect ed conspecifics occurs visually ................................ ................................ ............................... 29 2 4 Flume and quadrants used in the determining of t he range of PaV1 avoidance ................ 30 2 5 Geometric formation of artificial shelters used to determine the population structuring influence of PaV1 infected conspecific avoidance ................................ ......... 31 3 1 The sheltering choice (%) of focal lobsters in Y maze experiments used to determine the olfactory detectio n of PaV1 infected conspecifics ................................ ...................... 40 3 2 The sheltering choice (%) of focal lobsters in Y maze experiments used to determine the visual detectio n of PaV1 infected conspecifics ................................ .......................... 40 3 3 The amount of time spent in each location rela tive to health status in the PaV1 infected conspec ific avoidance range experiment ................................ ............................. 41 3 4 The percentage of lobsters remaining on the field sites over time ................................ .... 41 3 5 The mean distance moved away from central stimulus between the different factors irrespective of the other factors in PaV1 pop ulat ion structuring experiments ................... 42 3 6 The mean distance moved away from central stimulus between the different factors in PaV1 pop ulation structuring experiments ................................ ................................ ...... 43
8 LIST OF ABBREVIATION S ATP A denosine triphosphate CL C arapace length PaV1 Panulirus argus Virus 1 RhPV Rhopalosiphum padi virus s.d. Standard deviation
9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DETECTION OF PAV1 INFECTED CONSPECIFICS AND ITS POPULATION STRUCTURING DYNAMICS IN THE CARIBBEAN SPINY LOBSTER ( PANULIRUS ARGUS ) By Joshua Russ Anderson August 2011 Chair: Donald C. Behringer Major: Fisheries and Aquatic Science s PaV1 ( Panulirus argus virus 1) is a lethal contact transmitted pathogen that infects the Caribbean spiny lobster Panulirus argus Juvenile lobsters are highly susceptible to this pathogen, which prior to mortality causes fundamental changes in their ecology. However, P. argus can reduce infection risk by avoiding shelters inhabited by infected lobsters. Based on its role in many other aspects of lobster ecology such as conspecific attraction and m ate searching, I hypothesized that olfaction was the most likely mechanism by which lobsters detect PaV1 infected conspecifics A series of Y maze experiments were used to test this hypothesis and determine the source of the olfactory cue. Shelter avoidanc e behavior also has the potential to alter population structur ing dynamics and based on the type of cue, could be affected by local hydrodynamics. To investigate this, shelters and diseased cues were manipulated for wild populations in both high and low f low environments. Results showed that diseased avoidance is driven by olfaction via urine release, and moreover, the olfactory cue alone was equivalent in effectiveness to having a diseased lobster present and visible. When given a choice between shelteri ng with diseased or non diseased conspecifics, lobsters rarely sheltered with diseased individuals. In the field, detection of PaV1 infected conspecifics can alter structur ing dynamics
10 of natural populations especially under low flow regimes by inducing e migration and redistribution of non diseased animals relative to diseased conspecifics. In shelter limited environments, as occurs in many regions of the Caribbean, the unavailability of shelters due to disease avoidance could increase juvenile mortality a nd potentially impact future adult stocks.
11 CHAPTER 1 INTRODUCTION Olfaction is a sense widely used by organisms to understand their environment (Ache and Young 2005). The efficiency of olfactory mediated interactions is relative to the transmission and r ange of chemical detection, which in the terrestrial environment is limited by chemical volatility (Liley 1982) and bioenergetic cost of chemical production (Wilson 1975). In terrestrial organisms, olfaction is vital for reproduction (Averhoff and Richard son 1974, Carter et al. 1980), predation prey interactions (Apfelbach et al. 2005, Rosell and Sanda 2006), and interactions between organisms (Hurst 1990, Moore 1997, Bonadonna and Nevitt 2004). Compared to aquatic olfaction, terrestrial olfaction is well understood, with hundreds of chemical signals described for terrestrial arthropods (Herberholz 2007). Although better understood in the terrestrial environment, olfaction can dominate the sensory realm in the aquatic environment (Ache and Young 2005). In aquatic environments, the efficiency and effectiveness of chemical transmission is determined by the ability of a chemical to dissolve (solubility) and therefore disperse within its environment (Liley 1982) C hemical s can be r eleased from numerous ana tomical sources such as the slime covering of fish (Wilson 1975, Liley 1982), gill membranes (Liley 1982), and through urine (Bamber and Naylor 1997). Aquatic organisms employ olfaction to find food (Carr and Derby 1986, Finelli et al. 2000), migration/h oming (Hasler and Wisby 1951, Herrnkind 1969, Barbin et al. 1998), predator avoidance (Liley 1982, McCormick and Larson 2007), and intraspecific communication (Ameyaw Akumfi and Hazlett 1975, Stacey 2003, Diaz and Theil 2004). For example, food searching in the California spiny lobster Panulirus interruptus, is driven by the chemoreception of adenosine triphosphate (ATP ) in the water column (Zimmer Faust 1993). Likewise, homing by the black rockfish Sebastes inermis, observed by their strong attraction to their habitat after
12 displacement, is accomplished through olfactory pathways (Mitamura et al. 2005). Olfactory mediated predator avoidance is observed in the goby Asterropteryx semipunctatus, which can learn and associate novel chemical signals with d anger (Larson and McCormick 2005). Olfaction can also be found in mate searching by the female rock shrimp Rhynchocinetes typus, which is mediated predominantly through olfaction with females initially attracted to chemical signals emitted by males (Diaz and Theil 2004). C hemical communication is used by many decapods for a wide range of social interactions (Herberholz 2007). Among decapods, the source of the olfactory cue is often found in the urine (Atema 1986, Bushmann 1999, Breithaupt and Eger 2002 ) that contains pheromones used to mediate conspecific interactions (Wilson 1965, Bamber and Naylor 1997, Hor ner et al. 2008). Common urine mediated intraspecific interactions among decapods include: aggregation and homing (Childress and Herrnkind 1997, N evitt et al. 2000, Bushmann and Atema 1997), social interactions (Zimmer Faust et al. 1985, Bushmann 1999, Shabani et al. 2009), individual recognition (Karavanich and Atema 1998 a,b), reproduction (Atema and Engstrom 1971, Bamber and Naylor 1997, Kamio et al. 2008), and possibly disease detection (Behringer et al. 2006). The widespread use of urine in common interactions is credited to its rich soluble protein content, which increases its dispersal, and the ability of decapods to use appendage generated c urrents (Caldwell 1992, Gleeson et al. 1993, Kamio et al. 2008). urine could be employed in disease detection Moreover its metabolic waste concentration can be used to assess the internal and overall health of an organism (Breithaupt and Eger 2002, Atema and Steinbach 2007). Like individual recognition in many crustaceans, (Karavanich and Atema 1998 a,b, Hemsworth et al. 2007) viral detection could lead to an olfactory imprint of disease causing altered behaviors (Zulandt Schneider et al. 2001).
13 Crustacean s detect c hemical signals in the aquatic environment through chemoreceptors (Laverack 1988, Cate and Derby 2002) located primarily on the maxillipeds, pereiopods antennules, and antennae; however, crustaceans can have chemosensory cells on almost any part of the body (Atema and Steinbach 2007, Herberholz 2007). Maxillipeds, pere i opods, and other portions of the body are used for food tasting through tactile processes (Atema and Steinbach 2007). The antennules, conside red the main sensory organs in many crustaceans, have the largest proportion of chemosensory cells and are used for assessing the ir environment (Horner et al. 2008). When detecting odor plumes the antennules are commonly flicked to create flow that disrup ts the chemical boundary layer around the sensory cells, allowing a new signal to be processed (Schmitt and Ache 1979, Goldman and Patek 2002, Daniel et al. 2008). Olfaction can also be a population structuring mechanism, used by many crustaceans to for m conspecific aggregations (Bushmann and Atema 1997, Childress and Herrnkind 1997, Ritz 2000). Conspecific a ggregations are important because they d ecrease predation rates (Zimmer rovide reproductive refuge (Bushmann and Atema 1997, Atema and Steinbach 2007), and reduce energetic costs in swimming crustaceans (Ritz 2000). For example, the spiny lobster Panulirus interruptus uses olfaction to sense conspecific olfactory cues when searching for shelter aggregations (Zimmer Faust et al. 1985). Likewise, the American lobster Homarus americanus will form shelter aggregations in which the male protects the newly molted female during reproduction (Bushmann and Atema 1997). Many facto rs influence the formation of shelter aggregations including the number of suitabl e shelters, conspecific density and predation rates (Zimmer Faust et al. 1985, Nevitt et al. 2000, Behringer and Butler 2006).
14 While chemosensory cues are important in so cial interactions, visual cues have also evolved in organisms that inhabit shallow, well lit environments because they can be exchanged rapidly and efficiently from a distance (George and Main 1966, Acquistapace et al. 2002, Baldwin and Johnsen 2009). Vis ual communication is used in intraspecific interactions as found in female Lake Victoria cichlids Haplochromis nyererei complex, whose mate choice depends on coloration and display rates of suitable males (Seehausen and van Alphen 1998). Likewise, Baldwi n and Johnsen (2009) found that mate choice of the blue crab Callinectes sapidus can be stimulated solely by visual cues with males being attracted to females with red dactyls. Another example of the importance of visual cues lies in the diurnal migration of spiny lobsters wherein they align themselves in a straight formation via visual cues (Herrnkind 1969). Combined vi sual and chemical signals ( signal enhancement) are also commonly required in nature to induce behavioral response s because they increas e the efficiency of signal transmission (Smith 1969, Acquistapace et al. 2002, Herberholz 2007). Bimodal communication is employed in intraspecific interactions such as the male European freshwater crayfish, Austropotamobius pallipes (Acquistapace et al. 2002) and in the male snapping shrimp, Alpheus heterochaelis (Hughes 1996). T he female southern rock lobster, Jasus edwardsii, uses a combination of tactile, olfactory, and visual senses to distinguish between potential reproductive partners (Raethke et al. 2004). Signal enhancement plays a role in intraspecific interactions but also could play a vital role in forming shelter aggregations and population structuring due to its widespread use Regardless of the cue driving aggregation or population struct ur e it was long assumed that increased disease prevalence was a cost of sociality because social interactions can increase the spread of contact transmitted disease (Freeland 1979, Wright and Gompper 2005).
15 However, we now know that might not necessaril y be true. The ability to detect and avoid diseased conspecifics can reduce disease transmission by affecting population structure (Behringer et al. 2006, Ban et al. 2008). Although it has been shown that disease avoidance can affect the population struc ture of terrestrial and aquatic organisms (Thompson et al. 2002, Ban et al. 2008), only anecdotal data exists of these effects in the marine environment (Sousa 1991, Thompson et al. 2005, Behringer et al. 2006). For instance in the aquatic realm, the west ern mosquitofish Gambusa affinis (Tobler and Schlupp 2008), and the aphid Rhopalosiphum padi (Ban et al. 2008), avoid infected conspecifics creating segregation within the population. In the marine realm in which crustaceans shelter gregariously, the a voidance of shelters inhabited by infected conspecifics (Behringer et al. 2006) effectively removes the use of that shelter from those available for habitation, which in turn could have population spatial structuring affects if local disease prevalence is high. In marine environments with variable flow velocity, the efficiency of olfaction can be altered (Kleerekoper 1967, Weissburg and Zimmer Faust 1993, Finelli 2000). For instance, Finelli (2000) demonstrated that as flow velocity increases, odor plum e concentration decreases and physical structures ( e.g., sponges, octocoral, seagrass) dow nstream of the odor source turn the single odor plume into many small, dilute plumes. Another example is found in many teleost fish which cannot locate the source of stagnant olfactory cues, but can locate odor source when odor plumes are coupled with water flow (Kleerekoper 1967). Conversely, the efficiency of olfaction in blue crabs in high flow environments is decreased due to increased turbulence around the anten nule boundary layer (Weissburg and Zimmer Faust 1993) Furthermore if disease detection occurs via olfaction, local hydrodynamics could influence the area of avoidance.
16 The Caribbean spiny lobster Panulirus argus (Latrielle 1804), is one example of a decapod that employs visual and olfactory signals in an environment with variable flow. This gregarious animal is attracted to shelters inhabited by conspe cific s by using an aggregation odor This subsequently decreases predation by enabling conspecific s to find suitable shelter three times faster (Zimmer Faust et al. 1985, Childress and Herrnkind 1997, Childress and Herrnkind 2001). This olfactory driven behavior can help decrease predation on these vulnerable animals which are preyed upon by a suite o f predators (Smith and Herrnkind 1992, Mintz et al. 1994, Schratwieser 1999). O ther daily intraspecific interactions are also commonly mediated by olfaction (feeding Derby an d Cate 2001, predator avoidance Berger and Butler 2001 aggressive behaviors Shabani et al. 2 009) and visual cues (migration Herrnkind 1969). Their social nature implies that they should be highly susceptible to contact transmitted diseases, but Behringer et al. (2006) demonstrated that healthy P. argus can detect and avoid c onspecifics infected with the lethal virus Panulirus argus virus 1 (PaV1). However, infected individuals remain attracted to other conspecifics (non diseased or diseased) and will actively try to aggregate with them (Behringer et al. 2006). PaV1 is a con tact transmitted pathogen that prior to mortality causes fundamental changes in lobsters (Shields and Behringer 2004, Butler et al. 2008). Such fundamental changes include lethargy, tissue ischemia, depletion of energy reserves, and mortality (Shields and Behringer 2004). Although PaV1 can infect lobsters of all sizes, juveniles are most susceptible to this pathogen, succumbing more quickly than adults (mortality within 90 days) (Shields and Behringer 2004, Butler et al. 2008). Although it is known that lobsters can detect PaV1 infected conspecifics the mode of detection source of the cue and effects on population structuring dynamics are unknown Possible implications of disease avoidance in a shelter limited habitat could be emigration and redistrib ution away from
17 infected conspecifics. Using the Caribbean spiny lobster as a model species I tested the implic ations of disease avoidance on ju venile population structuring dynamics Hypothes e s: Determine the primary sense (visual versus olfactory) used in detecting PaV1 infected lobsters, and the source of the olfactory cue. Ho 1 : Caribbean spiny lobsters do not use chemical or visual cues to detect PaV1 infected conspeci fics H A1 : Detection of PaV1 infected spiny lobster s by visibly healthy lobsters occurs via chemical and visual signals. Ho 2 : If olfaction is critical to detecting conspecific cues relative to PaV1 in spiny lobster s, urine does not play a factor H A2 : If o lfaction is critical to detecting conspecific cues relative to PaV1 in spiny lobster s urine is the source of the olfactory cue Determine the range of infected lobster avoidance by non infected lobsters. Ho 1 : PaV1 infected lobster avoidance non infect ed lobster display, is not associate d with distance from cue source H A1 : PaV1 infected lobster avoidance non infected lobster display, is associated with the distance from the cue source. Evaluate how PaV1 affects juvenile population structure in the wild Ho 1 : PaV1 avoidance does not alter the spatial structure of wild spiny lobster populations H A1 : PaV1 avoidance alters the spatial structure of wild spiny lobster populations Evaluate the effect of flow on the spatial structuring function of the PaV1 avoidance behavior.
18 Ho 1 : Spatial avoidance of cues from PaV1 diseased lobster by non diseased lobsters is not positively associated with water flow rate H A1 : Spatial avoidance of cues from PaV1 diseased lobster by non diseased lobsters is positivel y associated with water flow rate.
19 CHAPTER 2 METHODS Mode of PaV1 infected Conspecific Detection and Cue Source Experimental Lobsters During the summer of 2010, juvenile Caribbean spiny lobsters 25 50 mm carapace length (CL) were hand collect ed from hard bottom habitat (< 3 m depth ) in Florida Bay, Florida (Figure 2 1) Lobsters were held in UV treated flow through tanks under natural photoperiod Lobsters were fed shrimp ad libitum every other day. Of these lobsters, 30 were inoculat ed with PaV1 by in jecting 0.1 mL of hemolymph from a visually PaV1 infected lobster ( Shields and Behringer 2004 ) through the membrane between the basis and the ischium of the fifth walking leg (Behringer et al. 2008). Inoculated animals were held separately from non diseas ed animals in recirculating tanks to ensure they were in a similar infection state. This inoculation method typically produces a 95% infection rate in a laboratory setting (Butler et al. 2008). Experimental Setup To determi ne the mode of PaV1 detection, Y maze experiments (Rebach 1996, Ratchford and Eggleston 1998, Diaz and Thiel 2004) were conducted in the summer fall 2010 Four 94cm long x 62cm wide x 20 cm tall Y mazes (79 L ) with a 72 cm long x 18cm tall central divider were constructed out of 1.3 cm p lywood coated with waterproof epoxy (Figure 2 2 ). To attract lobsters, each upstream arm of the Y maze held a conditioned artificial shelter (2 stacked concrete bricks; 20cm x 10cm x 5 cm) (Butler and Herrnkind 1997, Behringer and Butler 2006). Through th e center of the shelters, unfil tered, ambient seawater (25 1 C and 35 1 ppt) was gravity fed from isolated 121 L head tanks at a rate of 4.00 mL /s. From the shelters, water flowed downstream at a rate of 1 cm/s to an 18 cm tall, 1.9 cm diameter stand pi pe Dye tests
20 confirmed that water from each Y maze arm was unidirectional and restricted to its respective side before converging at the base of the Y maze Chemical Detection of PaV1 infected Conspecifics All Y maze experiments were conducted under natu ral photoperiod fo r an experimental duration of 8 h ending 1 h post dawn (when they are actively searching for shelters) In all olfactory Y maze trials 2 lobsters (specific treatments discussed below) were placed in a randomly assigned head tank 4 h prior to the start of the trial to infuse the water with their chemical signal. At the onset of the trial a non diseased (visually assured Shields and Behringer 2004 ) lobster was measured (CL; to the nearest 0.1 mm), s exed, and positioned at the base of the Y maze behind a porous screen. Following a 5 min period to reduce handling induced stress on the focal animal the screen was removed and the lobster was allowed to move about freely for the duration of the trial P reliminary observations indicated that lobsters (n = 12) sampled and thus received cues from both branches of the Y maze Eight hours into each trial, the shelter and associated treatment that the focal lobster selected were recorded. Between trials the setup was drained, flushed with fresh water three times (Diaz and Thiel 2004) an d allowed to air dry for 10 h To determine if P. argus can chemically detect and avoid PaV1 infected conspecifics, 3 olfactory experiments were conducted. The first experim ent tested the olfactory cues emitted from non diseased lobsters in one head tank versus olfactory cues emitted from seawater in the second head tank (n = 30). This experiment provided a baseline on the sheltering behaviors of focal animals and was expect ed to depict the gregarious nature of non diseased conspecifics. The second experiment tested the olfactory cues emitted from diseased lobsters in one head tank were tested against olfactory cues emitted from seawater in the second head tank (n = 20). In contrast to the previous experiment, this experiment provided information on the avoidance of
21 PaV1 infected individuals Finally, to determine if lobsters favored non diseased conspecifics over diseased conspecifics, olfactory cues emitted from non diseas ed lobsters in one head tank were tested against the olfactory cues emitted from a head tank that held diseased lobsters (n = 21). This experiment coupled the natural gregarity of focal animals to non diseased conspecifics with their natural avoidance of diseased conspecifics to determine their sheltering choice. In combination, t hese experiments provided information as to whether lobsters merely avoided diseased lobsters or rather were attracted to non diseased conspecifics. Crustaceans do not typically d emonstrate a side preference (Rebach 1996) or directional bias (Zimmer Faust 1993) during Y maze experiments, so these factors were ignored. PaV1 infected Conspecific Urine Detection To specifically test whether PaV1 infected conspecifics were detected c hemically through urine, Y maze experiments were conducted (n = 30) with the nephropores of diseased lobsters blocked (Bushmann 1999, Zulandt Schneider et al. 2001). The nephropores of the PaV1 diseased lobsters were blocked by blotting the nephropores dr y and then covering them with 2 coats of cyanoacrylate gel (Krazy Glue), followed by a thorough inspection to insure the complete coverage of the gel plug To reduce experimental bias, non diseased control lobsters were removed from their holding tank, b lotted dry (no glue applied) and held out of water for the same time period as treatment lobsters. Non diseased animals did not have their nephropores blocked because the purpose of the experiment was to see if blocking the nephropores of diseased animals in effect made them similar to non diseased animals. Observation s of the gel plug illustrated that it could be removed post experiment without detriment to the lobster. After application of the gel plug treatment lobsters ( nephropore blocked diseased b locked) were placed in one head tank while non diseased lobsters (unblocked) were placed in the other head tank 4 h prior to each trial If the PaV1 avoidance cue occurred as a consequence of urine emitted
22 from the nephropores, then the blocked nephropore lobsters would expected to be chemically The preceding Y maze experiments were used as controls because they served as a baseline of lobster sheltering behavior, providing focal lobster choice on shelt ers emitting olfactory cues of diseased or non diseased animals. Visual Cues in PaV1 Detection To determine if lobsters could visually detect PaV1 infected conspecifics Y maze s were modified by in stalling a sheet of glass (18 cm tall x 28cm wide) 6 c m in front of the each upstream sect ion of the Y maze (Figure 2 3 ) secured with silicone caulk ing to ensure that water was not exchanged between the Y maze and the partitioned section. Within this par titione d section two bricks (20cm x 10cm x 5 cm) standin g on end were placed to either side of the partition to keep the treatment lobster in the middle of the Y maze branch and in view of the focal lobster. The procedure was the same as described above except that no lobsters were present in the head tanks an d the water was released downstream of the glass partition. This experiment included three treatments: non diseased lobster visual cues versus seawater only (n = 30), disease lobster visual cues versus seawater only (n = 23), and non diseased lobster visua l cues versus diseased lobster visual cues (n = 30). Data Analyses To determine if the sheltering choice deviated from random, a two tailed binomial with the probability of sheltering set at 50% (random sheltering) was used. The treatments were considered independent and the sheltering probability stayed the same throughout the experiment (McClave and Sinc ich 2009). The power of the test (1 would correctly reject the null hypothesis (sheltering probability = 0.5). The value of the type II deviated from th e null hypotheses (0.5)
23 (McClave and Sincich 2009). Significance was determined at = 0.05 for this and all other experiments as described below PaV1 infected Conspecific Avoidance Range Experimental Setup and Procedure To determine the range at which P aV1 infected conspecifics elicits an avoidance reaction, a flume tank experiment was conducted. Previous studies with American lobster, Homarus americanus indicate d that they lose olfactory memory of previously encountered individuals after being separat ed for 2wks (Karavanich and Atema 1998 a,b). Similar observations have been observed in the common crayfish Cherax destructor (Hemsworth et al. 2007) and rusty crayfish Orconectes rusticus (Zulandt Schneider et al. 2001). Accordingly diseased lobster s in the present study were isolated f rom non diseased lobsters for 4 wks prior to the experiment to eliminate the chance of any olfactory recognition. Trials were conducted at night in a flume tank with a flow rate of 1.34 cm/s This flow rate was used b ecause based on the flow velocity at field sites it was a flow rate that would no t create excessive turbulence that would disrupt the odor plume. The bottom of the tank was divided into 3 quadra n ts (25cm long x 28 cm wide ) (Figure 2 4) At the downstrea m end of the experimental arena a porous screen was placed to keep lobsters from sheltering behind the stand pipe. The upstream end of the experiment al arena contained a brick shelter used to attract the focal lobsters to that end of the flume. Water was delivered to the shelter end of the tank via a 15 cm long PVC manifold. This manifold created a unidirectional flow and prevented focal lobsters from going beyond the shelter. To start the trial a treatment lobster (non diseased n = 19; diseased n = 29) was tethered to the center of the artificial shelter using a monofilament harness secured to the carapace with cyanoacrylate gel (Krazy Glue). A metal swivel connected between the carapace and tethering
24 line allowed the lobster to move about freely withi n the artificial shelter but prevented their movement into the experimental arena. A size matched non diseased lobster ( 5 mm CL of tethered animal) was then positioned at the d ownstream end of the tank for 5 min behind a porous screen to reduce handling induced stress after which the focal lobster was allowed to move about freely for 1 h. Diseased treatment animals were used in multiple trials due to the lack of diseased animals. T rials were recorded with a digital video camera under a red light for sub sequent analysis. Red lighting falls at the far end of the maximum spectral sensitivity of crustaceans (> 700nm wavelength) therefore minimizing any visual stimuli that might alter focal animal behavior (Cronin 1986). Between each trial the flume was drai ned, cleaned, and flushed with freshwater for 1h. Data Analysis Video of the trials was used to determine the amount of time spent sheltered or within each quadra n t ( shelter, 0 50 cm, 50 100 cm, and 100 150 cm). The mean amount of time spent in each quadr a n t was compared using an ANOVA with distance (sheltered, 0 50 cm, 50 100 cm, and 100 150 cm) and treatment (non diseased or diseased) as factors. ANOVA assumptions were assessed by comparing standardized residual plots ( residuals divided by the data stan dard deviation ; s.d. ) for this and other experiments as described below Influence of PaV1 Avoidance on Juvenile Population Structur ing Dynamics Experimental Setup In fall 2010, a tagging experiment was conducted in Florida Bay (USA) to assess the effe cts of PaV1 infected conspecific avoidance on spatial population structuring dynamics (Figure 2 1) The tagging experiment was conducted in artificial shelter arrays designed to take into account olfactory capabilities of P. argus by having multiple dista nces from the cue source (discussed below) Such capabilities include exchanging social signals over short distances, < 60
25 cm (Shabani et al. 2009), and orienting to odor plumes 2 m away (Derby et al. 2001). Eight artificial shelter arrays (Figure 2 5 ), w ere placed 100m a part in hard bottom habitat (~2 m depth) typically occupied by juvenile lobsters ( Herrnkind et al. 1997 ) Only large areas (at least 4 m 2 ) devoid of natural shelters were selected as sites for array placement. Artificial shelters were dou b le stacked concrete blocks (40cm x 20cm x 10cm) conditioned for 3 mo. in the ocean prior to use (Behringer and Butler 2006). Although the main focus of the study was on population structuring associated with PaV1 infected conspecific avoidance, I also soug ht to determine the importance of flow rate on and structur ing effect observed Therefore, 4 arrays were placed in low flow environments and 4 arrays placed in a high flow environments. An average flow (cm/s) (n = 3 for each flow regime) for the differing regimes was calculated using a digital / mechanical f low m eter (G.O. Environmental, Model 2030) set at the sit e for 24 h to record mean daily flow for each site and a t test was employed to determine significant differences between the flow regimes. Experi mental Procedure After colonization of the shelters (at least 3 lobsters per array), tagging surveys were conducted. Prior to dusk, 2 scuba divers used hand nets to capture each lobster on the site, record their CL, shelter location, mark it with a unique color coded antenna tag, and return it to its respective shelter. After the initial surveys (t 0 ) a visibly non diseased or diseased juvenile collected from nearby was tethered as described above with a 2 0 cm piece of monofilament line to the central bloc k in the shelter array. S ubseque nt surveys were conducted at 24 h (t 24 ) and 48 h (t 48 ) to allow untethered lobsters time to redistribute in response to the tethered treatment During the t 24 and t 48 surveys, divers recorded the location and tag code for ea ch tagged lobster. New arrivals at t 24 were treated as those at t 0 If tagged animals were encountered during the t 24 survey, the antennae tag was observed without actually capturing the lobster to reduce stress
26 induced movement. To understand the effec ts of stress caused by divers on lobster sheltering behavior, an initial survey (n = 6) was conducted 24 h before t 0 surveys. This pre trial survey used the same methods described above except no animal was tethered to the central shelter after the survey. This provided information on the amount of emigration from the site caused by the tagging process. Data Analysis The geometric formation of the shelters allowed multiple distances to be calculated from the central stimulus (Berger and Butler 2001). Thr ee measurements were taken from this formation: (1) lobsters 0.5 m from central shelter, (2) lobster s 1 m from the cent ral shelter, and (3) lobsters 2 m from the central shelter. Two separate repeated measures ANOVAs were used to reduce the variability assoc iated with time with the response variable log transformed Optimum power transformation was found by using a Box Cox transformation within SAS that finds the power transformation that maximizes reduction in model variability. Separate analyses were perf ormed due to unequal time factors between the two respective response variables ; the first analysis (response = count) had three entries per site ( t 0 t 24 t 48 ) while the second (response = distance moved) had two data points per site ( between t 0 t 24 and between t 24 t 48 ). Conclusions from both analyses were taken from the Type III tests of fixed effects. Significant ANOVAs were followed up with Tukey Kramer multiple comparison tests to determine the source of differences (McClave and Sincich 2009). M ultiple comparisons tests compared marginal means whi ch are means where the sum of square errors is minimized in the overall solution decreasing variability between the comparisons (McClave and Sincich 2009). The second analysis was used to determine if the movement of lobsters changed over time (redistribution) when a treatment animal was present Net movement was determined by adding the movement towards the treatment and subtracting movement away from the treatment
27 of each individual lobster on the s ite (lateral movement = 0). Emigration and immigration was subtracted or added to the total, respectively, relative to the abandoned/inhabited shelter distan ce from the central treatment. Immigration to shelters closes t to the central treatment (0.5 m) was scored as +2.5 because the immigrating lobster had to traverse closer to the central shelter Immigration to shelters 1 m from the central treatm ent was scored a +2. Like wise, immigration to shelters 2 m from the central treatment was scored a +1. Emigra tion was scored inversely to immigration; i.e. i f the abandoned shelter was 0.5 m from the central treatment than it was scored a 2.5.
28 Figure 2 1. Florida Bay, FL (USA), the location of lobster collection and study sites. Area of study sites are den Image credit: Florida Museum of Natural History ) Figure 2 2. Y maze design used to determine the mode of PaV1 infected conspecific detection the start of trials
29 Figure 2 3. Y maze set up for experiments used to determine if detection of PaV1 infected conspecifics occurs visually Hatch marked section represents a glass partition
30 Figure 2 4. Flume and quadra n ts used in the d etermining of the range of PaV1 avoidance
31 Figure 2 5. Geometric formation of artificial shelters used to determine the populati on structuring influence of PaV1 infected conspecific avoidance
32 CHAPTER 3 RESULTS Mode of PaV1 infected Conspe cific Detection and Cue Source Non diseased treatment animals in the Y maze experiment s ranged in size from 25.0 49.0 mm CL with a mean of 34.1 6.3 mm CL s.d. and diseased treatment animals ranged in size from 22.3 49.0 mm CL with a mean of 31.3 5. 1 mm CL s.d Non diseased treatment animals were significantly larger than diseased treatment animals (t test: df = 412.41; t value = 4.98; P < 0 .0001) but because diseased animals were significantly smaller (less urine output of smaller individuals; Shab ani et al. 2009) any bias would have been in favor of not finding a difference between disease and non diseased control treatments. Focal lobsters ranged in size from 22.1 49.8 mm CL with a mean of 33.4 6.3 mm CL s.d Results of the olfactory cue exp eriments are summarized in Table 3 1. Focal animals in the non diseased versus seawater only treatment selected the shelter with non diseased olfactory cue 67% of the time, but did not differ significantly from random sheltering (P = 0.0987; 1 ) (Figure 3 1). Focal lobsters in the diseased versus seawater only treatment preferred to shelter with no conspecific odor over diseased odor with sheltering probability significantly different than random (P = 0.0414; 1 ). Focal lob sters in the non diseased versus diseased treatments selected shelters with non diseased olfactory cues 86% of the time (P = 0.0015; 1 = 0.932). When the nephropores of diseased lobsters were blocked, focal lobsters selected shelters supplied with water from non diseased or diseased head tanks equally (P = 1.0000; 1 In no case was sheltering affected by treatments in visual cue experiments (Table 3 1 ). PaV1 infected Conspecific Avoidance Range ANOVA assumption checking indicated that s tan dardized residuals were independent of one another and had approximately equal variances. Although standardized residuals did not
33 conform to a normal distributed (Shapiro Wilk normality test: P < 0.0001), the data was considered normally distributed due t o the general robustness of ANOVAs in determining significant differences in means even if the data are not normally distributed. During the flume tank experiment to determine the range of PaV1 infected conspecifics by focal lobsters treatment lob sters ra nged in size from 25.1 39.2 mm CL with a mean of 32.2 4.1 mm CL s.d Focal lobsters ranged in size from 23.6 38.1 mm CL with a mean of 30.4 3.7 mm CL s.d Results indicate that the presence of a diseased or non diseased animal alone was not signi ficant (P = 0.9986), but dist ance from disease presence (P < 0.0001) and interaction between distance and disease presence differed significantly (P = 0.0002) (Table 3 2 ). The distances 0 50 cm (P = 0.2393) and 50 100 cm (P = 0.6351) did not differ si gnificantly between treatments, but 100 150 cm (P = 0.0352) and within shelter (P = 0.0002) differed significantly (Figure 3 2 ). Influence of PaV1 Avoidance on Juvenile Population Structuring Dynamics ANOVA assumptions checking indicated that all assum ptions were met, the standardized residuals were found to be normally distributed, independent of one another, and had approximately equal variances. A Shapiro Wilk normality test among the standardized residuals indicated that they conformed to a normal distribution (count analysis: P = 0.7780; movement analysis: P = 0.1686). In 28 experimental trials (disease n = 12, non disease n = 16 or high flow n = 14, low flow n = 14) 258 natural focal lobsters were caught and tagg ed, ranging in size from 10.0 65 .1 mm CL with a mean of 29.6 7.5 mm CL s.d There was a mean of 8.1 4.8 wild lobsters per site at t 0 surveys. In the16 non diseased (control) experiments, tethered lob sters ranged in size from 29.5 49.8 mm CL with a mean of 37.0 6.4 mm CL s.d. a nd in the 12 disease experiments tethered lobsters ranged in size from 24. 3 37.1 mm CL with a mean of 29.9 3.9
34 mm CL s.d Carapace length of tethered animals differed significantly between treatments ( t test: df = 25.25 ; t value = 3.01; P = 0.00 58 ), but because diseased animals were significantly smaller (less urine output of smaller individuals ; Shabani et al. 2009) any bias would have been in favor of not finding a difference between disease and non diseased control treatments. Flow at low flow sit es (0.20 cm/s; 0.74 s.d. ) was significantly less (df = 2.02; t value = 4.79; P = 0.0403) than flow at high flow sites (7.94 cm/s; 2.79 s.d. ). Repeated measures ANOVA analysis indicated no significant difference between pretrial and t 0 wild lobster cou nts on the sites (df = 9; F = 0.00; P = 0.9665), no r were numbers affected by flow regime (df = 22; F = 0.04; P = 0.8509) or distance (df = 22; F = 0.08; P = 0.9278). Trial surveys comparing the changes in lobster count over time indicated that health s tatus (P = 0.0009) and time (P < 0.0001) differed significantly (Table 3 3 ) (Figure 3 4). Multiple comparison analysis of the marginal means was used to determine that within the diseased treatment the number of lobsters at low flow sites was not signific antly different tha n the number of lobsters at the high flow sites (P = 0.0810). Within the low flow treatment, the re were significantly fewer lobsters present at diseased sites (P = 0.0010). Non diseased treatment sites did not differ significantly betw een flow rates (P = 0.9989) nor did health status differ significantly und er high flow rates (P = 0.7408) (Table 3 4 ). Within health status, the number of lobsters located within shelters at different distances from the central shelter also differed sig nificantly. The number of lobsters at 0.5 m from the central shelter differed significantly between health status (P = 0.0314) wh ile the number of lobsters at 1m (P = 0.0690) and 2 m (P = 0.1929) from the central shelter did not differ significantly (Table 3 5) At the distance of 0.5 m there were fewer lobsters located at the diseased sites compared to the non diseased sites. Within the diseased or non diseased
35 treatments there were no significant differences between the distances (Table 3 5 ). M ultiple co mparison analysis of the interaction of heath status, flow rate, and distance from central shelter rev ealed that at a distance of 0.5 m in a low flow regime there were significantly fewer lobsters on disease treatment sites (df = 146; t v alue = 3.95; Adj P = 0.0066). Repeated measures ANOVA results indicated that mean distance moved by lobsters differed significantly as a consequence of health status (P < 0.0001 ) and dista nce from central treatment (P < 0.0001) but flow regime was not a sign ificant factor (P = 0. 8331 ) (Table 3 6 ). Within the diseased treatment the movement did not differ significantly between low and high flow rate (P = 0. 6741 ) nor did flow rate differ significantly within the non diseased treatment (P = 0.2 091 ). W ithin the high flow treatment movement did not differ significantly between the diseased and non diseased treatment (P = 0. 3389 ). Within the low flow treatment, lobsters mov ed significantly further away from the central treatment shelter on diseased treatment sit es (P = 0.00 03 ) (Table 3 7 ) (Figure 3 6). W ithin health status, lobster movement was significantly differe nt among the distance from the central treatment (Table 3 8 ). Within the diseased treatment, movement differed significantly when comparing the origi nal inhabited shelter distance fr om the central treatment of 0.5m to 1 m (P < 0.0001 ) and 0.5m to 2 m (P < 0.0001). On the other hand, when comparing the original inhabited shelter distance from the central treatment, 1m did not d iffer significantly from 2 m (P= 0.0 506 ). Within the original in habited shelter distance of 0.5 m from the central shelter lobsters moved significantly father a way from the central treatment at diseased sites (P < 0.0001 ) There were no significant differences in lobster movement fro m shelters located 1 m (P = 0. 8325 ) and 2 m (P = 0. 9651 ) from the central treatment
36 Table 3 1. Results determining the detection of PaV1 infected conspecific s (a) R esults from the olfactory treatments only. (b ) R esults from the visual treatments only Experiment n/n P 1 a. Olfaction treatments Non diseased/s eawater only 20/10 0.0987 0.432 Diseased/s eawater only 05/15 0.0414 0.617 Diseased/n on diseased 03/18 0.0015 0.932 Diseased blocked/n on diseased 15/15 1.0000 0.043 b. Visual treatments Non diseased/s eawater only 15/15 1.0000 0.043 Diseased/s eawater only 14/09 0.4049 0.143 Diseased/n on diseased 14/16 0.8555 0.057 Note: Treatments were tested against random sheltering (50%) P indicate a significant result. Table 3 2 Results used to determine the range of the PaV1 olfactory avoidance cue (a) Overall ANOVA results. (b) M u ltiple comparison test compari ng the distance relative to non diseased and diseased treatments a. Source df Typ e III SS F value P Treatment 1 1.5 0.00 0.9986 Distance 3 256 985 190 .0 186.18 < 0.0001 Treatment x d istance 3 9 667 823 .0 6.79 0.0002 b. Multiple comparison t est Within shelter 1 6 761 731 .0 14.25 0.000 2 0 50 cm from shelter 1 661 570 .0 1.39 0.2393 50 100 cm from shelter 1 107 237 .0 0.23 0.6351 100 150 cm from shelter 1 2 137 287 .0 4.50 0.0352 Note: P
37 Table 3 3 Results used to determine effect of disease avoidance on the number of lobsters per site over time and under high and low flow regimes Source df F value P Time 71 20.17 < 0.0001 Health status 146 11.42 0.0009 Flow 146 2.95 0.0879 Distance from stimulus 146 1.19 0.3069 Health status X flow 146 3.63 0.0586 Healthy status X time 146 2.24 0.1097 Flow X time 146 0.84 0.4345 Health status x distance from stimulus 146 0.26 0.7734 Flow X dis tance from stimulus 146 3.60 0.0298 Time X health s tatus X distance from stimulus 146 2.28 0.0248 Healthy status X flow rate X distance from stimulus 146 2.69 0.0715 Note: P Table 3 4 Results used to determine the spatial structuring capabilities of PaV1 relative to the number of lobsters on the site Comparison w ithin Comparison b etween df Difference of means SE t value Adj P Dise ase Flow regime 146 0.24 0.101 2.40 0.0810 Non diseased Flow regime 146 0.01 0.088 0.14 0.9989 Low flow Health status 146 0.35 0.092 3.84 0.0010 High flow Health status 146 0.10 0.097 1.02 0.7408 Comparison Disease Low Non diseased high 146 0.34 0.088 3.88 0.0009 Disease High Non diseased low 146 0.11 0.101 1.10 0.6894 Note: This multiple comparison is of health status and flow rate P indicate a significant result.
38 Table 3 5 Resul ts used to determine the spatial structuring capabilities of PaV1 relative to the number of lobsters on the site Comparison w ithin Comparison b etween df Difference of means SE t value Adj P Disease 0.5m 1 .0 m 146 0.02 0.064 0.34 0.9994 0.5m 2 .0 m 146 0.09 0.064 1.44 0.7056 1 .0 m 2 .0 m 146 0.07 0.064 1.10 0.8815 Non diseased 0.5m 1 .0 m 146 0.00 0.057 0.04 1.0000 0.5 m 2 .0 m 146 0.03 0.057 0.54 0.9942 1 .0 m 2 .0 m 146 0.03 0.057 0.59 0.9918 0.5 m Health status 146 0.25 0.083 3.05 0.0314 1 m Health status 146 0.23 0.083 2.77 0.0690 2m Health status 146 0.19 0.083 2.32 0.1929 Note: This multiple comparison analysis is of health status and distance from the central shelter P va Table 3 6 Results used to determine the influence of disease avoidance on the juvenile populations structure Source d f F value P Time 43 2.92 0.0 946 Health sta tus 290 17.82 < 0.0001 Flow 290 0.04 0. 8331 Distance from stimulus 290 91.30 < 0.0001 Health status X flow 290 4.20 0.0414 Healthy status X time 290 1.44 0.2315 Flow X time 290 0.34 0.5629 Health status X distance from stimulus 290 9.71 < 0.0001 Flow X distance from stimulus 290 0 .54 0.5812 Time X health status X distance from stimulus 290 1.06 0.3777 Healthy status X flow rate X distance from stimulus 290 2.15 0.1186 Note: This analysis was used to determine differences in lob ster movement over time P values
39 T able 3 7 Results used to determine the affect of PaV1 infected conspecific avoidance on lobster movement Comparison w ithin Comparison b etween df Difference of means SE t value Adj P Disease Flow regime 290 0. 25 0.2 23 1. 13 0. 6741 Non diseased Flow regime 290 0.3 1 0.1 57 1. 95 0.2 091 Low flow Health status 290 0.8 5 0.2 06 4.14 0.00 03 High flow Health status 290 0. 30 0.178 1.67 0.3389 Comparison Disease l ow Non diseased high 290 0.5 5 0.204 2.68 0.0383 Disease h igh Non diseased low 290 0. 60 0.181 3.34 0.0 0 52 Note: This multiple comparison analysis is of health status and flow rate P values denoted with Table 3 8 Results used to determine the affect of PaV1 infected conspecific avoidance on lobster movement Comparison w ithin Comparison b etween df D ifferences of means SE t value Adj P Disease 0.5m 1 .0 m 290 1.86 0.2 52 7.37 < 0.00 01 0.5m 2 .0 m 290 2.46 0.2 40 10.25 < 0.0001 1 .0 m 2 .0 m 290 0. 60 0.2 11 2.86 0.0506 Non diseased 0.5m 1 .0 m 290 0.79 0. 150 5.26 < 0. 000 1 0.5m 2 .0 m 29 0 1. 30 0.1 4 0 9.02 < 0.0001 1 .0 m 2 .0 m 290 0. 51 0. 152 3.34 0.0 119 0.5 m Health status 290 1.32 0. 231 5. 71 < 0.00 01 1 .0 m Health status 290 0. 25 0.2 08 1. 21 0. 8325 2 .0 m Health status 290 0.16 0. 191 0 .8 1 0. 9651 Note: This multiple compa rison analysis is of health status and distance from the central shelter P
40 Figure 3 1. The sheltering choice (%) of focal lobsters in Y maze experiments used to determine the olfactory detection of PaV1 infected conspecifics The horizontal dotted line repr esents random sheltering (50%). Within the hash marks, paired bars with an were significantly different from random sheltering Error bars represent 1 standard error Figure 3 2. The shel tering choice (%) of focal lobsters in Y maze experiments used to determine the visual detection of PaV1 infected conspecifics The horizontal dotted line represents random sheltering (50%). Within the hash marks, no treatments were significantly differen t from random sheltering Error bars represent 1 standard error
41 Figure 3 3. The amount of time spent in each location relative to health status in the PaV1 infected conspecific avoidance range experiment. Paired bars signifi cant ly di fferent from one another Error bars represent 1 standard error Figure 3 4. The percentage of lobsters remaining on the field sites over time between non diseased and diseased treatments in the PaV1 population structuring experiments. Standard error bars significant differences
42 Figure 3 5. The mean distance moved away from central stimulus between the different factors irrespective of the other factors in PaV1 population structuring experiments Paired bars within the hash m Error bars represent 1 standard error
43 Figure 3 6. The mean distance moved away from central stimulus between the different factors in PaV1 population structuring experiments A bar wi diffe rent than the other treatment. Error bars represent 1 standard error
44 CHAPTER 4 DISCUSSION The avoidance behavior that visibly healthy P. argus displays in response to PaV1 infected conspecifics is driven by olfactory cues p resent in the urine of diseased individuals D etection of PaV1 infected conspecifics led to spatial struc turing of juvenile populations on experimental field sites through, emigration and redistribution. Compounding the affects of disease avoidance on th e juvenile population is the influence of differing flow regimes on the detection of PaV1 infected conspecifics with greater emigration in low flow environments. E ffects of emigration and redistribution in shelter limited habitats could increase juvenile mortality (Behringer and Butler 2010) and subsequently induce a population bottleneck by potentially increasing PaV1 transmission or predation. Olfaction is a highly efficient mode of habitat assessment in the marine environment used during feeding (Carr and Derby 1986, Finelli et al. 2000), migration/homing events (Herrnkind 1969, Barbin et al. 1998, Mitamura et al. 2005), predator avoidance (Liley 1982, McCormick and Larson 2007), and other species specific interactions (Ameyaw Akumfi and Hazlett 1975, Diaz and Theil 2004, Kamio et al. 2008). This efficiency holds true with disease detection as non diseased lobster cues were more favorable than diseased cues to animals searching for shelter. T he source of the avoidance cue was determined to be urine be cause when the nephropores of infected lobsters were blocked the avoidance behavior ceased. To focal animals searching for suitable shelters, the olfactory cues of diseased animals with blocked nephropores mirrored non diseased animals. Like the formation of shelter aggregations, many social interactions are also chemically mediated including individual recognition (Karavanich and Atema 1998 a,b, Zulandt Schneider et al. 2001) and reproduction (Atema and Engstrom 1971, Diaz and Theil 2004, Kamio et al. 200 8). The actual cue being detected in the urine is unknown, although it could be
45 the metabolic waste concentration (Breithaupt and Eger 2002, Atema and Steinbach 2007) or the actual physical PaV1 vir u s, future work is needed to determine this avoidance dr iving cue. On the other hand, visual cues do not appear to play a role in PaV1 avoidance. O lfactory detection likely obfuscates the need for visual detection and bimodal signal enhancement was not evident Olfactory detection of PaV1 infected conspecifi cs might also dominate because it is sensitive to infected conspecifics even when they are not yet visibly infected or infectious (Behringer et al. 2006). Although healthy lobsters avoided infected conspecifics, the range of the avoidance response in a la boratory setting could not be determined During avoidance range experiments, focal lobsters spent significa ntly more time further away (+1 m) from diseased animals than non diseased animals, but also spent a considerable amount of time sheltered with dise ased and non diseased conspecifics Compared to PaV1 infected conspecifics detection experiments, the range detection experiment gave focal animals only one shelter choice, of which 86% of focal animals sheltered with the diseased treatment for a period o f time. Although animals instinctively avoid diseased individuals, under these experimental conditions it appears that shelter limitation, and presumably the threat of predation, outweighed the threat of disease transmission. Behringer and Butler (2010) found similar sheltering behaviors of P. argus within a laboratory setting with focal animals sheltering with diseased animals when only one shelter was provided. While the findings reported herein are possibly due to the inadequate length of the flume, t he fact that animals spent significantly more time further away from the diseased treatment than non diseased treatment, mirrors previous P. argus olfactory experiments in which lobsters could exchange social signals over short distances and orient to odor plumes 2 m away (Derby et al. 2001, Shabani et al. 2009). Other factors that could have influence the outcome of this
46 experiment was the fact that turbulence caused by the flow could have disrupted odor plumes decreasing their readability and the trial t ime (1h) may have been too short to display altered sheltering behaviors. Field experiments demonstrated that disease avoidance dramatically altered the spatial structur ing of juvenile populations via redistribution and emigration. Not only did diseased t reatment sites have fewer lobsters over time, but the remaining animals moved farther away from the diseased lobster A fter a diseased conspecific finds its way into the population it would be advantageous for neighboring lobsters to quickly move away. A lthough the maximum distance was not determined in the laboratory, in the field diseased lobsters had an effect on wild juveniles at least 1 m away. These results are consistent with experiments indicating that urine is diluted to 0.1% of its original stre ngth 1 m from the source which suggests that specific urine driven behaviors probably only function over short distances (Shabani et al. 2009) compared to non conspecific driven behavior experiments in which juvenile lobsters were able to chemically detect and avoid octopus predators from up to 2m away (Berger and Butler 2001). Other factor s that could have played a role in the animal movement and the reduction of animals at the sites could be the stress caused by handling ( induced by divers collecting and tagging animals ) and also predation which can be very high on these vulnerable juveniles (Smith and Herrnkind 1992). During surveys no predators were noticed on the sites although they could still be present, but because focal animals at the diseased and non diseased treatment sites experience the same handling and predation rates, the different in the number of lobsters at the sites can be attributed to the experimental treatments. I nfluence of disease on population structuring dynamics was affected by fl ow with high flow regimes displaying a lesser effect on movement patterns and emigration. In low flow
47 environments the presence of a diseased animal significantly altered population structure by inducing emigration, possibly due to the concentrated disea sed odor plume diffusing over the habitat without being diluted by turbulence (Finelli 20 0 0). High velocity flow and physical structures downstream of the odor source (concrete shelters) eliminated the affects of disease avoidance, presumably by diluting the odor plume produced by the diseased lobster (Finelli 20 0 0). Reduced olfactory efficiency in high velocity flow could possibly result from turbulence antenn ule boundary layer before it can be read (Weissburg and Zimmer Faust 1993). In areas of daily tidal fluctuations, periods of turbulence could possibly render odor plumes unreadable, but during changes in tidal direction animals could have the opportunity to sense the odor plume and respond. The small scale population structuring effects of the avoidance of PaV1 infected conspecifics have similarities to terrestrial and freshwater disease avoidance. For example, black spot disease Uvulifer sp., can affect the population structure of the western mosquitofish Gambusia affinis (Tobler and Schlupp 2008). Both healthy and dise ased females avoid shoaling with infected conspecifics (Tobler and Schlupp 2008). Similarly, a study on the effects of Rhopalosiphum p adi virus (RhPV) on the aphid R. padi demonstrated that healthy aphids were attracted to uninfected conspecific aggregations but not attracted to infected aggregations; spatially segregating infected and healthy individuals on host plants (Ban et al. 2008 ). Infected aphids were also f ound to be more sensitive to a species specific alarm cue causing them to leave the host plant resulting i n increased mortality and a possible decrease in transmission (Ban et al. 2008). Organisms such as R. padi and P. arg us avoid infected conspecifics through olfactory processes, causing segregation within the population. Segregation is apt to take a heavy toll on
48 diseased individuals as they have been shown to suffer higher levels of predation, probably due to their let hargic condition (Ban et al. 2008, Behringer and Butler 2010). T hese examples illustrate that the population structuring affects of disease avoidance is apparent in a broad suite of environments. Higher levels of predation possibly resulting from disea se avoidance could result in a population bottleneck of juveniles in shelter limited habitats when PaV1 prevalence is high (Herr nkind et al. 1997, Behringer and Butler 2006, Behringer and Butler 2010). In habitats that are shelter limited naturally or due to habitat degradation (Butler and Herrnkind 1997, Herr nkind et al. 1997), the avoidance of infected animals and the subsequent redistribution and emigration could result in increased predation. As juveniles move away from diseased conspecifics to areas of unknown shelter density or across open substrate their predation risk is likely to increase, especially among these mostly small, vulnerable juveniles (Eggleston et al. 1990, Eggleston et al. 1992, Smith and Herrnkind 1992). The vulnerability of unshel tered juveniles is due to them having a large suite of suitable predators (Eggleston et al. 1990, Smith and Herrnkind 1992, Mintz et al. 1994) and the lack of protection that group defense offer s (Zimmer Faust and Spanier 1987, Eggleston et al. 1990, Smith and Herrnkind 1992). Likewise, non emigrating animals also could have a risk of increased mortality due to the increased risk of contracting PaV1 from nearby infected conspecifics. Such situations of high disease prevalence and limited shelters could ca use juveniles to incur increased mortality due to predation or disease transmission. It would seem maladaptive that disease avoidance would cause an increase in predation on healthy lobsters but the overall increase in fitness gained by avoiding disease may be greater
49 While disease avoidance would seem logical some organisms are indifferent or even attracted to diseased conspecifics (Han et al. 2008, Bouwman and Hawley 2010). For example, two amphibians Bufo boreas and Rana cascadae were indifferent to conspecifics infected with the pathogen Batrachochytrium dendrobatidis, and healthy R. cascadae were actually found more often than not aggregated with infected individuals (Han et al. 2008). Bouwman and Hawley (2010) studied the effects of the pathoge n Mycoplasma gallisepticum on the male house finch Carpodacus mexicanus and its reaction towards infected conspecifics. Healthy males actually preferred to feed next to overtly infected finches because they were less aggressive and therefore less energe tically costly to healthy males; i.e. the energetic cost of feeding near a lethargic infected male was lower (Bouwman and Hawley 2010). Although cases of disease attraction do exist, it is usually induced by an attraction to lethargic conspecifics and th erefore less energetically costly to the healthy individual (Bouwman and Hawley 2010). Although few studies have dealt with the hypothetical implications of parasites or diseases on marine food webs or population structure (Sousa 1991, Thompson et al. 20 05, Behringer and Butler 2010), no previous studies have determined the influence of disease avoidance on the spatial structure of a marine population. Pathogens can disrupt nearly every reproduction, feeding, social inte ractions) and through avoidance have the ability to influence population structur ing A number of studies have dealt with disease avoidance and the possible implications on population or community structure (Thompson et al. 2002, Behringer et al. 2006, Ba n et al. 2008). Community structure (e. g. host aggregation) can influence the transmission of contact transmitted pathogens (Shields and Behringer 2004, Wright and Gompper 2005) but host movement due to emigration, disease avoidance, or resource depleti on can influence this transmission (Freeland 1979, Wright and Gompper 2005). By
50 manipulating resources Wright and Gompper (2005) found that the prevalence and species richness of endoparasites within wild raccoons Procyon lotor increased as the host con tact rate increased around clumped resources. Not only does the ability to detect infectious pathogens influence the community structure but infected individual behavior can also play a role (Behringer et al. 2006, Ban et al. 2008, Han et al. 2008). S imilar to the Caribbean spiny lobster (Behringer et al. 2006), infected individuals of some organisms are still inherently attracted to their conspecifics (Ban et al. 2008, Han et al. 2008). For example the amphibian B. boreas infected with the fungal pa thogen Batrachochytrium dendrobatidis, is still attracted to healthy individuals and will actively try to aggregate with them (Han et al. 2008). Likewise, infected individuals of this study species will shelter with infected and healthy animals alike (B ehringer et al. 2006). This could possibly result in the aggregation of infected individuals within the population and therefore a decrease in transmission (possible disease sink) but could also result in diseased animals moving into shelters inhabited by non diseased individuals and transmitting the disease to these individuals. The behavioral response of diseased individuals to diseased olfactory cues needs to be further studied to fully understand how their attraction and movement towards healthy indiv iduals can influence population structuring dynamics Conclusion: The ability of visibly healthy lobsters to detect and avoid PaV1 infected conspecifics is driven by chemosensory cues and this can influence the spatial structure of wild juvenile populati on s through redistribution and emigration However, local hydrodynamics can alter this effect, presumably by affecting the distribution and detection of odor plumes. Although effective at decreasing disease transmission disease avoidance could also incr ease vulnerab ility to predation, further altering population structure. Clearly disease can play a p ivotal role in
51 animal behaviors and in concert with the physical environment, can be a major structuring force for populations, both demographically and sp atially.
52 LIST OF REFERENCES Ache, B. W., J. M. Young 2005. Olfaction: D iverse species, conserve d principles. Neuron 48:417 430. Acquistapace, P. L. Aquiloni, B. A. Hazlett F. Gherardi 2002. Multimodal communication in crayfish: S ex recognition duri ng mate search by male Austropotamobius pallipes. Canadian Journal of Zoology 80:2041 2045. Ameyaw Akumfi, C., and B. A. Hazlett. 1975. Sex recognition in the crayfish Procambarus Clarkii. Science 190:1225 1226. Apfelbach, R, C. D. Blanchard, R. J. Blanc hard, R. W. Hayes, and I. S. McGregor. 2005. The effects of predator odors in mammalian prey species: A review of field and laboratory studies. Neuroscience & Biobehavorial Reviews 29:1123 1144. Atema, J. 1986. Review of sexual selection and chemical comm unication in the lobster Homarus americanus Canadian Journal of Fisheries and Aquatic Science 43:2283 2390. Atema, J., and D. G. Engstrom. 1971. Sex pheromone in the lobster, Homarus americanus. Nature 232:261 263. Atema, J., and M. A. Steinbach. 2007 Chemical communication and social behavior of the lobster Homarus americanus and other decapods crustacean. Pages 115 144 in : J. E. Duffy and M. Thiel editors. Evolutionary ecology of social and sexual systems: Crustaceans as model organisms. Oxford Univ ersity Press, New Yo rk, New York. Averhoff, W. W., and R. H. Richardson. 1974. Pheromonal control of mating patterns in Drosophila melanogaster. Behavior Genetics 4:207 225. Baldwin, J., and S. Johnsen. 2009. The importance of color in mate choice of the blue crab Callinectes sapidus Journal of Experimental Biology 212:3762 3768. Bamber, S. D., and E. Naylor. 1997. Sites of release of putative sex pheromone and sexual behavior in female Carinus maenas (Crustacea: Decapoda). Estuarine, Coastal and Shelf Science 44:195 202. Ban, L., E. Ahmed, V. Ninkovic, G. Delp, and R. Glinwood. 2008. Infection with an insect virus affects olfactory behaviour and interactions with host plant and natural enemies in an aphid. Entomologia Experimentalis et Applicata 127:10 8 117. Barbin, G. P., S. J. Parker, and J. D. MeCleave. 1998. Olfactory clues play a critical role in estuarine migration of silver phase American eels. Environmental Biology of Fishes 53:283 291.
53 Behringer, D. C., and M. J. Butler IV. 2006. Density depe ndent population dynamics in juvenile Panulirus argus (Latreille): T he impact of artificial density enhancement. Journal of Experimental Marine Biology and Ecology 334:84 95. Behringer, D. C., and M. J. Butler IV. 2010. Disease avoidance influences shelte r use and predation in Caribbean spiny lobster. Behavioral Ecology and Sociobiology 64:747 755. Behringer, D. C., M. J. Butler IV, and J. D. Shields. 2006. Avoidance of disease by social lobsters. Nature 441:421. Behringer, D. C., M. J. Butler IV, and J. D. Shields. 2008. Ecological and physiological effects of PaV1 infection on the Caribbean spiny lobster ( Panulirus argus Latreille). Journal of Experimental Marine Biology and Ecology 359:26 33. Berger, D. K., and M. J. Butler IV. 2001. Octopuses influen ce den selection by juvenile Caribbean spiny lobster. Marine and Freshwater Research 52:1049 1053. Bonadonna, F., and G. A. Nevitt. 2004. Partner specific odor recognition in an Antarctic seabird. Science 306:835. Bouwman, K. M., and D. M. Hawley. 2010. Sickness behaviour acting as an evolutionary trap? Male house finches preferentially feed near diseased conspecifics. Biology Letters 6:462 465. Breithaupt. T., and P. Eger. 200 2. Urine makes the difference: C hemical communication in fighting crayfish mad e visible. The Journal of Experimental Biology 205:1221 1231. Bushmann, P. J. 1999. Concurrent signals and behavioral plasticity in blue crab ( Callinectes sapidus Rathbun) courtship. Biological Bulletin 197:63 71. Bushmann, P. J., and J. Atema. 1997. She lter sharing and chemical courtship signals in the lobster Homarus americanus Canadian Journal of Fisheries and Aquatic Science 54:647 654. Butler IV, M. J., D. C. Behringer, and J. D. Shields. 2008. Transmission of Panulirus argus virus 1 (PaV1) and its effects on the survival of juvenile Caribbean spiny lobsters. Disease of Aquatic Organisms 79:173 182. Butler IV, M. J., and W. F. Herrnkind. 1997. A test of recruitment limitation and the potential for artificial enhancement of spiny lobster ( Panulirus argus ) populations in Florida. Canadian Journal of Fisheries and Aquatic Science 54:452 463. Caldwell, R. L. 1992. Recognition, signaling and reduced aggression between former mates in a stomatopod. Animal Behaviour 44:11 19.
54 Carr, W., and C. D. Derby. 1 986. Behavioral chemoattractants for the shrimp, Palaemonetes pugio identification of active components in food extracts and evidence of synergistic mixture interactions. Chemical Senses 11:49 64. Carter, C. S., L. L. Getz, L. Gavish, J. L. McDermott, a nd P. Arnold. 1980. Male related pheormones and activation of female reproduction in the prairie vole ( Microtus ochrogaster ). Biology of Reproduction 23:1038 1045. Cate, H. S., and C. D. Derby. 200 2. Hooded sensilla homologues: S tructural variations of a widely distributed bimodal chemomechanosensillum. Journal of Comparative Neurology. 444:345 357. Childress, M. J., and W. F. Herrnkind. 1997. Den sharing by juvenile Caribbean spiny lobsters ( Panulirus argus ) in nursery habitat: C ooperation or coincidence ? Marine and Freshwater Research 48:751 758. Childress, M. J., and W. F. Herrnkind. 2001. The guide effect influence on the gregariousness of juvenile Caribbean spiny lobsters. Animal Behaviour 62:465 472. Cronin, T. W. 1986. Photoreception in marine inv ertebrates. American Zoologist 26:403 415. Daniel, P. C., M. Fox, and S. Mehta. 2008. Identification of chemosensory sensilla mediating antennular flicking behavior in Panulirus argus the Caribbean spiny lobster. Biological Bulletin 215:24 33. Derby, C. D., P. Steullet, A. J. Horner, and H. S. Cate. 2001. The sensory basis of feeding behaviour in the Caribbean spiny lobster, Panulirus argus Marine and Freshwater Resources 52:1339 1350. Diaz, E. R., and M. Thiel. 2004. Chemical and visual communication d uring mate searching in rock shrimp. Biological Bulletin 206:134 143. Eggleston, D. B., and R. N. Lipcius. 1992. Shelter selection by spiny lobster under variable predation risk, social conditions, and shelter size. Ecology 73:992 1011. Eggleston, D. B., R. N. Lipcius, D. L. Miller, and L. Coba Cetina. 1990. Shelter scaling regulates survival of juvenile Caribbean spiny lobster Panulirus argus. Marine Ecology Progress Series 62:79 88. Finelli, C. M. 2000. Velocity and concentration distributions in turbu lent odor plumes in the presence of vegetation mimics: A flume study. Marine Ecology Progress Series 207:297 309. Finelli, C. M., N. D. Pentcheff, R. K. Zimmer, and D. S. Wethey. 2000. Physical Constraints on ecological processes: A field test of odor med iated foraging. Ecology 83:784 797.
55 Freeland, W. J. 1979. Primate social groups as biological islands. Ecology 60:719 728. George, R. W., and A. R. Main. 1966. The evolution of spiny lobsters (Palinuridae): A study of evolution in the marine environment. Evolution 21:803 820. Gleeson, R. A., W. Carr, and H. G. Trapido Rosenthall. 1993. Morphological characteristics facilitating stimulus access and removal in the olfactory organ of the spiny lobster, Panulirus argus insight form the design. Chemical Sen ses 18:67 75. Goldman, J. A., and S. N. Patek. 2002. Two sniffing strategies in Palinurid lobsters. The Journal of Experimental Biology 205:3891 3902. Han, B. A., P. W. Bradley, and A. R. Blaustein. 2008. Ancient behaviors of larval amphibians in respons e to an emerging fungal pathogen, Batrachochytrium dendrobatidis Behavioral Ecology and Sociobiology 63:241 250. Hasler, A. D., and W. J. Wisby. 1951. Discrimination of stream odors by fishes and its relation to parent stream behavior. The American Natur alist 135:223 238. Hemsworth, R., W. Villareal, B. W. Patullo, and D. L. Macmillan. 2007. Crustacean social behavioral changes in response to isolation. Biological Bulletin 213:187 195. Herberholz, J. 2007. The neural basis of communication in crustacean s. Pages 72 89 in : J. E. Duffy and M. Thiel editors. Evolutionary ecology of social and sexual systems: Crustaceans as model organisms. Oxford University Press, New York, New York Herrnkind, W, 1969. Queuing behavior of spiny lobsters. Science 164:1425 1 427. Herrnkind, W. F., M. J. Butler IV, J. H. Hunt, and M. Childress. 1 997. Role of physical refugia: I mplications from a mass sponge die off in a lobster nursery in Florida. Marine and Freshwater Research 48:759 769. Horner, A. J., S. P. Nickles, M. J. Weissburg, and C. D. Derby. 2006. Source and specificity of chemical cues mediating shelter preference of Caribbean spiny lobsters ( Panulirus argus ). The Biological Bulletin 211:128 140. Horner, A. J., M. J. Weissburg, and C. D. Derby. 2008. The olfactory pathway mediates sheltering behavior of Caribbean spiny lobsters, Panulirus argus to conspecific urine signals. Journal of Comparative Physiology A Neuroethology Sensory Neural and Behavioral Physiology 194:243 253. Hughes, M. 1996. The f unction of conc urrent signals: V isual and chemical communication in snapping shrimp. Animal Behaviour 52:247 257. Hurst, J. L. 1990. Urine marking in population of wild house mice Mus domesticus Rutty. I. Communication between males. Animal Behaviour 40:209 222.
56 Kamio, M., M. A. Reidenbach, and C. D. Derby. 2008. To paddle or not: C ontext dependent courtship display by male blue crabs, Callinectes sapidus The Journal of Experimental Biology 211:1243 1248. Karavanich, C., and J. Atema. 1998a. Olfactory recognition of u rine signals in dominance fights between male lobster, Homarus americanus Behaviour 135:719 730. Karavanich, C., and J. Atema. 1998b. Individual recognition and memory in lobster dominance. Animal Behaviour 56:1553 1560. Kleerekoper, H. 1967. Some aspec ts of olfaction in fishes, with special reference to orientation. American Zoologist 7:385 395. Larson, J. K., and M. I. McCormick. 2005. The role of chemical alarm signals in facilitating learned recognition of novel chemical cues in coral reef fish. Ani mal Behaviour 69:51 57. Laverack, M. S. 1988. The diversity of chemoreceptors. Pages 287 312 in : J. Atema, R. R. Fay, A. N. Popper, and W. N. Tavolga, editors. Sensory biology of aquatic animals. Springer Verlag, New York, New York Liley, N. R. 1982. Ch emical communication of fish. Canadian Journal of Fisheries and Aquatic Science 39:22 35. McClave, J. T., and T. Sincich. 2009. Statistics: Eleventh edition. D. Lynch editor. Pearson Prentice Hall Press, Upper Saddle River, New Jersey. McCormick, M. I., and J. K. Larson. 2007. Field verification of the use of chemical alarm cues in a coral reef fish. Coral Reefs 26:571 576. Mintz, J. D., R. N. Lipcius, D. B. Eggleston, and M. S. Seebo. 1994. Survival of juv enile Caribbean spiny lobster: E ffects of shelte r size, geographic location and conspecifics abundance. Marine Ecology Progress Series 112:255 266. Mitamura, H., N. Arai, W. Sakamoto, Y. Mitsunaga, H. Tanaka, Y. Mukai, K. Nakamura, M. Sasaki, and Y. Yoneda. 2005. Role of olfaction and vision in homing behavior of black rockfish Sebastes inermis Journal of Experimental Marine Biology and Ecology 322:123 134. Moore, A. J. 1997. The evolution of social signals: M orphological, functional, and genetic integration of the sex pheromone in Nauphoeta cinerea. Evolution 51:1920 1928. Nevitt, G., N. D. Pentcheff, K. J. Lohmann, and R. K. Zimmer. 2000. Den selection by the spiny lobster Panulirus argus : T esting attraction to conspecific odors in the field. Marine Ecology Progress Series 203:225 231.
57 P. and D. A. Ritz. 1988. Escape responses of gregarious mysids (Crustacea: Mysidacea): T owards a general classification of escape responses in aggregated crustaceans. Journal of Experimental Marine Biology and Ecology 116:257 272. Raethke, N., A. B. MacDi armid, and J. C. Montgomery. 2004. The role of olfaction during mating in the southern temperate spiny lobster Jasus edwardsii Hormones and Behavior 46:311 318. Ratchford, S. G., and D. B. Eggleston. 1998. Size and scale dependent chemical attraction co ntribute to an ontogenetic shift in sociality. Animal Behaviour 56:1027 1034. Rebach, S. 1996. Role of prey in food recognition by rock crabs, Cancer irroratus Say. Journal of Chemical Ecology 22:2197 2207. Ritz, D. A. 2000. Is social aggregation in aqua tic crustaceans a strategy to conserve energy? Canadian Journal of Fisheries and Aquatic Sciences 57:59 67. Rosell, F., and J. I. Sanda. 2006. Potential risks of olfactory signaling: T he effect of predators on scent marking by beavers. Behavioral Ecology 17:897 904. Schratwieser, J. 1999. The impact of resident and transient predators on the population dynamics of juvenile Caribbean spiny lobster ( Panulirus argus ) in Florida Bay, Florida. Thesis. Old Dominion University, Norfolk, Virginia Schmitt, B. C ., and B. W. Ache. 1979. Olfaction: R esponses of a decapods crustacean are enhanced by flicking. Science 205:204 206. Seehausen, O., and J. J. M. van Alphen. 1998. The effect of male coloration on female mate choice in closely related Lake Victoria cichli ds ( Haplochromis nyererei complex). Behavioral Ecology and Sociobiology 42:1 8. Shabani, S., M. Kamio, and C. D. Derby. 2009. Spiny lobsters use urine borne olfactory signaling and physical aggressive behaviors to influence social status of conspecifics. The Journal of Experimental Biology 212:2464 247 4 Shields, J. D., and D. C. Behringer. 2004. A new pathogenic virus in the Caribbean spiny lobster Panulirus argus from the Florida Keys. Diseases of Aquatic Organisms 59:108 11 8. Smith, K. N., and W. F. Herrnkind. 1992. Predation on early juvenile spiny lobsters Panulirus argus (Latreille): I nfluence of size and shelter. Journal of Experimental Marin e Biology and Ecology 157:3 18. Smith, W. J. 1969. Messages of vertebrate communication. Science 165:145 150.
58 Sousa, W. P. 1991. Can models of soft sediment community structure be complete without parasites? American Zoologist 31:821 830. Stacey, N. 2003. Hormones, pheromones and reproductive behavior. Fish Physiology and Biochemistry 28:229 235. Thompson, K G., R. B. Nehring, D. C. Bowden, and T. Wygant. 2002. Response of rainbow trout Oncorhynchus mykiss to exposure to Myxobolus cerebralis above and below a point source of infectivity in the upper Colorado River. Disease of Aquatic Organisms 49:171 178. T hompson, R. M., K. N. Mouritsen, and R. Poulin. 2005. Importance of parasites and their life cycle characteristics in determining the structure of a large marine food web. Journal of Animal Ecology 74:77 85. Tobler, M., and I. Schlupp. 2008. Influence of black spot disease on shoaling behaviour in female western mosquitofish, Gambusia affinis (Poeciliidae, Teleostei). Environmental Biology of Fishes 81:29 34. Weissburg, M. J., and R. K. Zimmer Faust. 1993. Li fe and death in moving fluids: H ydrodynamic eff ect on chemosensory mediated predation. Ecology 74:1428 1443. Wilson, E. O. 1965. Chemical communication in the social insects. Science 149:1064 1071. Wilson, E. O. 1975. Sociobiology: T he new synthesis. Harvard University Pr ess, Cambridge, Massachusetts Wright, A. N., and M. E. Gompper. 2005. Altered parasite assemblages in raccoons in response to manipulated resource availability. Oecologia 144:148 156. Zimmer Faust, R. K. 1993. ATP: A potent prey attractant evoking carnivory. Limnology and Oceanogr aphy 38:1271 1275. Zimmer Faust, R. K., and E. Spanier. 1987. Gregariousness an d sociality in spiny lobsters: I mplications for den habitation. Journal of Experimental Marine Biology and Ecology 105:57 71. Zimmer Faust, R. K., J. E. Tyre, and J. F. Case. 1985. Chemical attraction causing aggregation in the spiny lobster, Panulirus interruptus (Randall), and its probable ecological significance. Biological Bulletin 169:106 118. Zulandt Schneider, R. A., R. Huber, and P. A. Moore. 2001. Individual and status recognition in the crayfish, Orconectes rusticus : The effects of urine release on fight dynamics. Behaviour 138:137 153.
59 BIOGRAPHICAL SKETCH Josh grew up in a no stoplight town on the foothills of the Missouri Ozarks. He graduated from Russel lville High School with an avid interest in the outdoors which he cultivated through his early years growing up and working on the family farm. He furthered his education by attending the University of Missouri Columbia during which he was employed by t he Missouri Department of Conservation. At the Conservation Department, he worked with the endangered Niangua darter ( Etheostoma nianguae ) to determine the health of the population and also worked on understanding the impacts of stream erosion and human i nduced stream degradation on and is still a proud alumnus of the University of Missouri. His loathing of the bitter Missouri winters and his interest in the ec ology of marine systems helped him find his way to the sunny University of Florida degree with Dr. Donald Behringer in the Summer of 2011